20005 research outputs found
Sort by
Efficient numerical implementations for the Maxey-Riley-Gatignol Equation
No full textAs a second-order, implicit integro-differential equation with singular kernel at initial time, the resolution of the Maxey-Riley-Gatignol Equation (MRGE) presented many challenges. This thesis presents new Finite Difference methods (FDM) based on the reformulation stated in Prasath et al. (2019), where the MRGE is reformulated as a boundary condition of the 1D Heat equation. The performance of the FDM are compared against existing schemes for six flow fields, providing advice on when each method performs best. The influence of the Basset History Term is studied on particle trajectories, clusters and Lagrangian Coherent Structures, delivering guidelines on when it can be omitted.Als eine implizite Integro-Differentialgleichung zweiter Ordnung mit singulärem Kern zur Anfangszeit stellte die Lösung der Maxey-Riley-Gatignol-Gleichung (MRGE) viele Herausforderungen dar. Diese Dissertation stellt neue Finite-Differenzen-Methoden (FDM) vor, die auf der in Prasath et al. (2019) angegebenen Reformulierung basieren, bei der die MRGE als Randbedingung der 1D-Wärmeleitungsgleichung reformuliert wird. Die Performanz der FDM wird mit bestehenden Verfahren für sechs Strömungsfelder verglichen, wobei Ratschläge gegeben werden, wann jede Methode am besten funktioniert. Der Einfluss des Basset-Geschichtsterms wird auf Partikelbahnen, Cluster und Lagrangian Coherent Structures untersucht, wobei Richtlinien gegeben werden, wann dieser Term weggelassen werden kann.Deutsche Forschungsgemeinschaft (DFG
Improved solubility predictions in scCO₂ using thermodynamics-informed machine learning models
No full textAccurate solubility prediction in supercritical carbon dioxide (scCO2) is crucial for optimizing experimental design by eliminating unnecessary and costly trials at an early stage, thereby streamlining the workflow. A comprehensive solubility database containing 31975 records has been compiled, providing a foundation for developing predictive models applicable to a diverse class of chemical compounds, with a particular focus on drug-like substances. In this study, we propose a Domain-Aware Machine Learning approach that incorporates thermodynamic properties governing phase transitions to solubility predictions in scCO2. Predictive models were developed using the CatBoost algorithm and a graph-based architecture employing directed message passing to identify the most effective approach. Furthermore, auxiliary properties of the solute, including melting point, critical parameters, enthalpy of vaporization, and Gibbs free energy of solvation, were predicted as part of this work. The findings underscore the efficacy of incorporating domain-specific thermodynamic features to enhance the predictive accuracy of scCO2 solubility modeling. The interpretation and the applicability domain assessment have confirmed the qualitative selection of the employed descriptors, demonstrating their ability to generalize to unique compounds that fall outside the defined domain
Compost organic matter content varied five-fold and determined compost quality across 107 composts of the North Sea Region
No full textComposting is a widely used method to process organic waste residues. It results in a valuable product for soil application and use in growing media. The aim of this study was to investigate the variation in characteristics of composts produced in the North Sea Region, and the factors determining this variation. A total of 107 composts were categorized into two composting practices (produced on a farm or on a commercial composting facility) and three feedstock groups (manure combined with other wastes; green waste; fruit, vegetable and garden waste (fvg)), and measured for 67 physical, chemical and biological characteristics. Variation in the results was large, e.g., up to a factor 20 and 11 for total microbial biomass and potassium content, respectively, underlining the importance of compost characterization to target the intended compost use. Organic matter (OM) content varied between 14 and 73% of dry matter and was larger for Belgian composts compared to composts from The Netherlands, Denmark, Germany and Scotland. The OM content was positively correlated with total microbial biomass, cation exchange capacity and content of nitrogen (N) and phosphorus (P) of composts. Farm composts, irrespective of the OM effect, exhibited higher total microbial biomass compared to commercial composts. Compost prepared from green waste had lower N and P contents compared to compost prepared from fvg or manure waste. The study documents characteristics in composts from diverse composting practices and feedstocks, providing a benchmark and enabling targeted improvements as a first step towards tailormade compost
A machine learning approach for planning valve-sparing aortic root reconstruction
No full textAbstract Choosing the optimal prosthesis size and shape is a difficult task during surgical valve-sparing aortic root reconstruction. Hence, there is a need for surgery planning tools. Common surgery planning approaches try to model the mechanical behaviour of the aortic valve and its leaflets. However, these approaches suffer from inaccuracies due to unknown biomechanical properties and from a high computational complexity. In this paper, we present a new approach based on machine learning that avoids these problems. The valve geometry is described by geometrical features obtained from ultrasound images. We interpret the surgery planning as a learning problem, in which the features of the healthy valve are predicted from these of the dilated valve using support vector regression (SVR). Our first results indicate that a machine learning based surgery planning can be possible
Influence of different transposon families on genomic stability of Shewanella Oneidensis MR1
No full textShewanella oneidensis, recognised as an important model organism for exoelectrogenic electron transport, has been extensively studied for its potential applications in bioelectrochemical systems. To date, the activity of transposable elements in this organism has not been conclusively investigated. This study focused on transposases, specifically insertion sequences (IS), which make up approximately 4.7% of the organism's genome, and evaluated their impact on genome stability under stress conditions. Using whole genome sequencing, two IS families, ISSOD1 and ISSOD2, were identified as the most active, both showing similar transposition patterns across all tested stressors. A CRISPR/dCas9 cytosine deaminase system was used to introduce stop codons in the ISSOD2 transposase genes, resulting in a significant reduction of transposition events under stress conditions. Analysis of transposition patterns revealed a high frequency of insertions occurring on the megaplasmid, which predominantly carries non-essential genes. Experiments performed here to delete the megaplasmid resulted in the elimination of approximately 35% of its sequence, including an unexpected complete loss of the ori/repA region. Therefore, it was hypothesised that the megaplasmid either exists in a metastable state, possibly representing a cointegrated intermediate within the ISSOD9 (Tn3 member) transposition mechanism, or consists of two replicons that have been combined in previous assemblies due to long overlapping homologies resulting from the presence of ISSOD9. These findings highlight the dynamics of transposable elements in S. oneidensis and suggest strategies to improve strain stability by inactivating these elements and at least reducing megaplasmid sequences. Such approaches could improve the suitability of the organism for industrial applications
European Test Symposium Teams: an anniversary snapshot
No full textThe IEEE European Test Symposium (ETS) has been facilitating progress in electronic systems testing since its launch in 1996. On the occasion of its 30th anniversary, this collaborative paper gathers sections by 21 ETS teams to outline their influential ideas and milestones. Each team's section highlights historical perspective, current research, frameworks and projects as well as forward-looking research agendas in the area of electronic-based circuits and systems testing, reliability, safety, security and validation. This anniversary summary documents how research of various ETS teams, exemplifying the test community, has been evolving and transitioning from concepts to practical standards and Electronic Design Automation (EDA) tools and flows. This legacy is a strong base to drive the next generation of advances in electronic systems testing
Pull-off force prediction in viscoelastic adhesive Hertzian contact by physics augmented machine learning
No full textPredicting the adhesive properties of viscoelastic Hertzian contacts is crucial for diverse engineering applications, including robotics, biomechanics, and advanced material design. This study introduces a novel physics-augmented machine learning (PA-ML) framework as a hybrid approach to study the maximum adherence force of a Hertzian indenter unloaded from a viscoelastic substrate, bridging the gap between analytical models and data-driven solutions. The PA-ML model is capable of rapidly predicting the pull-off force in an Hertzian profile unloaded from a broad band viscoelastic material, with varying Tabor parameter, preload and retraction rate. Compared to previous models, the PA-ML approach provides fast yet accurate predictions in a wide range of conditions, properly predicting the effective surface energy and the work-to-pull-off. The integration of the analytical model provides critical guidance to the PA-ML framework, supporting physically consistent predictions. We demonstrate that physics augmentation enhances predictive accuracy, reducing mean squared error (MSE) while increasing model interpretability. We provide data-driven and PA-ML models for real-time predictions of the adherence force in soft materials like silicons and elastomers opening to the possibility to integrate PA-ML into materials and interface design. The models are openly available on Zenodo and GitHub
Data-driven probabilistic evaluation of logic properties with PAC-confidence on Mealy machines
No full textCyber-Physical Systems (CPS) are complex systems that require powerful models for tasks like verification, diagnosis, or debugging. Often, suitable models are not available and manual extraction is difficult. Data-driven approaches then provide a solution to, e.g., diagnosis tasks and verification problems based on data collected from the system. In this paper, we consider CPS with a discrete abstraction in the form of a Mealy machine. We propose a data-driven approach to determine the safety probability of the system on a finite horizon of n time steps. The approach is based on the Probably Approximately Correct (PAC) learning paradigm. Thus, we elaborate a connection between discrete logic and probabilistic reachability analysis of systems, especially providing an additional confidence on the determined probability. The learning process follows an active learning paradigm, where new learning data is sampled in a guided way after an initial learning set is collected. We validate the approach with a case study on an automated lane-keeping system
How to crack a SMILES: automatic crosschecked chemical structure resolution across multiple services using MoleculeResolver
No full textAbstract: Accurate chemical structure resolution from textual identifiers such as names and CAS RN® is critical for computational modeling in chemistry and related fields. This paper introduces MoleculeResolver, an automated, robust Python-based tool designed to address inconsistencies and inaccuracies commonly encountered when converting chemical identifiers to canonical SMILES strings. MoleculeResolver systematically crosschecks structures retrieved from multiple reputable chemical databases, implements rigorous identifier plausibility checks, standardizes molecular structures, and intelligently selects the most accurate representation based on a unique resolution algorithm. Scientific contribution: Benchmarks across diverse datasets confirm that MoleculeResolver significantly enhances precision, recall, and overall reliability compared to traditional single-source methods, proving its utility as a valuable resource for chemists, data scientists, and researchers engaged in high-quality molecular data analysis and predictive model development